This is an R Markdown document. Markdown is a simple formatting syntax for authoring HTML, PDF, and MS Word documents. For more details on using R Markdown see http://rmarkdown.rstudio.com.
When you click the Knit button a document will be generated that includes both content as well as the output of any embedded R code chunks within the document. You can embed an R code chunk like this:
library(readr)
diabetes_dataset <- read_csv("diabetes_prediction_dataset.csv")## Rows: 100000 Columns: 9
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (2): gender, smoking_history
## dbl (7): age, hypertension, heart_disease, bmi, HbA1c_level, blood_glucose_l...
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.You can also embed plots, for example:
diabetes_dataset## # A tibble: 100,000 × 9
## gender age hypertension heart_disease smoking_history bmi HbA1c_level
## <chr> <dbl> <dbl> <dbl> <chr> <dbl> <dbl>
## 1 Female 80 0 1 never 25.2 6.6
## 2 Female 54 0 0 No Info 27.3 6.6
## 3 Male 28 0 0 never 27.3 5.7
## 4 Female 36 0 0 current 23.4 5
## 5 Male 76 1 1 current 20.1 4.8
## 6 Female 20 0 0 never 27.3 6.6
## 7 Female 44 0 0 never 19.3 6.5
## 8 Female 79 0 0 No Info 23.9 5.7
## 9 Male 42 0 0 never 33.6 4.8
## 10 Female 32 0 0 never 27.3 5
## # ℹ 99,990 more rows
## # ℹ 2 more variables: blood_glucose_level <dbl>, diabetes <dbl># male datatset
male_data = diabetes_dataset %>% filter(gender == "Male")
# female dataset
female_data = diabetes_dataset %>% filter(gender == "Female")
female_data## # A tibble: 58,552 × 9
## gender age hypertension heart_disease smoking_history bmi HbA1c_level
## <chr> <dbl> <dbl> <dbl> <chr> <dbl> <dbl>
## 1 Female 80 0 1 never 25.2 6.6
## 2 Female 54 0 0 No Info 27.3 6.6
## 3 Female 36 0 0 current 23.4 5
## 4 Female 20 0 0 never 27.3 6.6
## 5 Female 44 0 0 never 19.3 6.5
## 6 Female 79 0 0 No Info 23.9 5.7
## 7 Female 32 0 0 never 27.3 5
## 8 Female 53 0 0 never 27.3 6.1
## 9 Female 54 0 0 former 54.7 6
## 10 Female 78 0 0 former 36.0 5
## # ℹ 58,542 more rows
## # ℹ 2 more variables: blood_glucose_level <dbl>, diabetes <dbl># males and females within original dataset that have a "normal" A1C
female_data %>% filter(HbA1c_level <= 5.7) %>% tally()## # A tibble: 1 × 1
## n
## <int>
## 1 27397male_data %>% filter(HbA1c_level <= 5.7) %>% tally()## # A tibble: 1 × 1
## n
## <int>
## 1 18865# count of people (male and female) with both heart disease and diabetes
diabetes_dataset %>% filter(diabetes == 1, heart_disease == 1)## # A tibble: 1,267 × 9
## gender age hypertension heart_disease smoking_history bmi HbA1c_level
## <chr> <dbl> <dbl> <dbl> <chr> <dbl> <dbl>
## 1 Male 67 0 1 not current 27.3 6.5
## 2 Male 57 1 1 not current 27.8 6.6
## 3 Male 80 0 1 former 24.4 7.5
## 4 Male 75 0 1 not current 28.1 7.5
## 5 Male 69 0 1 former 24.1 6.8
## 6 Female 59 0 1 never 60.3 8.8
## 7 Male 80 0 1 former 33.0 6
## 8 Female 62 1 1 former 44.2 8.2
## 9 Female 62 1 1 never 43.2 8.8
## 10 Female 76 0 1 former 25.7 9
## # ℹ 1,257 more rows
## # ℹ 2 more variables: blood_glucose_level <dbl>, diabetes <dbl># count of overweight people based on bmi who have heart disease
diabetes_dataset %>% group_by(bmi >= 30) %>% filter(heart_disease == 1)## # A tibble: 3,942 × 10
## # Groups: bmi >= 30 [2]
## gender age hypertension heart_disease smoking_history bmi HbA1c_level
## <chr> <dbl> <dbl> <dbl> <chr> <dbl> <dbl>
## 1 Female 80 0 1 never 25.2 6.6
## 2 Male 76 1 1 current 20.1 4.8
## 3 Female 72 0 1 former 27.9 6.5
## 4 Male 67 0 1 not current 27.3 6.5
## 5 Female 77 1 1 never 32.0 5
## 6 Female 59 0 1 ever 23.1 6.5
## 7 Male 68 1 1 current 27.3 5
## 8 Male 59 0 1 ever 30.8 5
## 9 Female 80 0 1 never 29.6 5.8
## 10 Male 57 1 1 not current 27.8 6.6
## # ℹ 3,932 more rows
## # ℹ 3 more variables: blood_glucose_level <dbl>, diabetes <dbl>,
## # `bmi >= 30` <lgl>diabetes_dataset %>% group_by(bmi >= 30) %>% filter(heart_disease == 1) ## # A tibble: 3,942 × 10
## # Groups: bmi >= 30 [2]
## gender age hypertension heart_disease smoking_history bmi HbA1c_level
## <chr> <dbl> <dbl> <dbl> <chr> <dbl> <dbl>
## 1 Female 80 0 1 never 25.2 6.6
## 2 Male 76 1 1 current 20.1 4.8
## 3 Female 72 0 1 former 27.9 6.5
## 4 Male 67 0 1 not current 27.3 6.5
## 5 Female 77 1 1 never 32.0 5
## 6 Female 59 0 1 ever 23.1 6.5
## 7 Male 68 1 1 current 27.3 5
## 8 Male 59 0 1 ever 30.8 5
## 9 Female 80 0 1 never 29.6 5.8
## 10 Male 57 1 1 not current 27.8 6.6
## # ℹ 3,932 more rows
## # ℹ 3 more variables: blood_glucose_level <dbl>, diabetes <dbl>,
## # `bmi >= 30` <lgl># "obese men" with bmi higher than 30 and that have diabetes (tally on second line)
male_data %>% filter(bmi >= 30, diabetes == 1)## # A tibble: 1,903 × 9
## gender age hypertension heart_disease smoking_history bmi HbA1c_level
## <chr> <dbl> <dbl> <dbl> <chr> <dbl> <dbl>
## 1 Male 50 0 0 former 37.2 9
## 2 Male 53 0 0 current 30.8 6.6
## 3 Male 76 0 0 never 31.9 7.5
## 4 Male 63 1 0 ever 35.1 5.8
## 5 Male 48 1 0 current 36.1 6.8
## 6 Male 37 0 0 never 37.2 7
## 7 Male 36 0 0 not current 46.1 6.2
## 8 Male 50 0 0 never 31.8 7.5
## 9 Male 43 0 0 never 69.4 7.5
## 10 Male 43 1 0 not current 40.9 6.6
## # ℹ 1,893 more rows
## # ℹ 2 more variables: blood_glucose_level <dbl>, diabetes <dbl>male_data %>% filter(bmi >= 30, diabetes == 1) %>% tally()## # A tibble: 1 × 1
## n
## <int>
## 1 1903# "obese women" with bmi higher than 30 and that have diabetes (tally on second line)
female_data %>% filter(bmi >= 30, diabetes == 1)## # A tibble: 2,330 × 9
## gender age hypertension heart_disease smoking_history bmi HbA1c_level
## <chr> <dbl> <dbl> <dbl> <chr> <dbl> <dbl>
## 1 Female 67 0 0 never 63.5 8.8
## 2 Female 36 0 0 current 32.3 6.2
## 3 Female 77 0 0 never 31.7 6.5
## 4 Female 47 0 0 never 36.5 7.5
## 5 Female 61 0 0 not current 39.4 9
## 6 Female 80 0 0 former 36.2 6.5
## 7 Female 52 1 0 never 50.3 6.6
## 8 Female 68 0 0 No Info 40.3 7.5
## 9 Female 70 0 0 not current 33.2 7.5
## 10 Female 67 0 0 former 32.3 7
## # ℹ 2,320 more rows
## # ℹ 2 more variables: blood_glucose_level <dbl>, diabetes <dbl>female_data %>% filter(bmi >= 30, diabetes == 1) %>% tally()## # A tibble: 1 × 1
## n
## <int>
## 1 2330# "underweight men" with bmi lower than 19 and that have diabetes (tally on second line)
male_data %>% filter(bmi <= 19, diabetes == 1)## # A tibble: 21 × 9
## gender age hypertension heart_disease smoking_history bmi HbA1c_level
## <chr> <dbl> <dbl> <dbl> <chr> <dbl> <dbl>
## 1 Male 42 0 0 current 11.9 6
## 2 Male 6 0 0 never 15.7 6.1
## 3 Male 71 1 0 former 13.2 6.6
## 4 Male 14 0 0 never 19.0 6.6
## 5 Male 54 0 0 never 18.9 6
## 6 Male 61 1 0 never 18.4 6.5
## 7 Male 4 0 0 never 18.7 6
## 8 Male 51 0 0 current 17.8 6.2
## 9 Male 80 1 0 current 19.0 6.6
## 10 Male 6 0 0 No Info 15.6 9
## # ℹ 11 more rows
## # ℹ 2 more variables: blood_glucose_level <dbl>, diabetes <dbl>male_data %>% filter(bmi <= 19, diabetes == 1) %>% tally()## # A tibble: 1 × 1
## n
## <int>
## 1 21# "underweight women" with bmi lower than 19 and that have diabetes (tally on second line)
female_data %>% filter(bmi <= 19, diabetes == 1)## # A tibble: 57 × 9
## gender age hypertension heart_disease smoking_history bmi HbA1c_level
## <chr> <dbl> <dbl> <dbl> <chr> <dbl> <dbl>
## 1 Female 79 0 0 not current 18.1 7
## 2 Female 4 0 0 No Info 15.0 6.5
## 3 Female 51 0 0 current 17.4 7
## 4 Female 9 0 0 never 16 6.1
## 5 Female 60 0 0 No Info 17.9 8.2
## 6 Female 13 0 0 No Info 17.3 6.2
## 7 Female 80 0 0 never 17.4 6.5
## 8 Female 8 0 0 No Info 14.3 7.5
## 9 Female 80 0 0 never 17.8 6.2
## 10 Female 78 1 0 not current 17.7 8.8
## # ℹ 47 more rows
## # ℹ 2 more variables: blood_glucose_level <dbl>, diabetes <dbl>female_data %>% filter(bmi <= 19, diabetes == 1) %>% tally()## # A tibble: 1 × 1
## n
## <int>
## 1 57# the assumption is that overweight people are more likely to have diabetes. Below is the code and tally of MEN who are overweight in terms of bmi and DONT have diabetes
male_data %>% filter(bmi >= 30, diabetes == 0)## # A tibble: 7,445 × 9
## gender age hypertension heart_disease smoking_history bmi HbA1c_level
## <chr> <dbl> <dbl> <dbl> <chr> <dbl> <dbl>
## 1 Male 42 0 0 never 33.6 4.8
## 2 Male 15 0 0 never 30.4 6.1
## 3 Male 40 0 0 current 36.4 6
## 4 Male 30 0 0 never 33.8 6.1
## 5 Male 34 0 0 never 31.2 5.8
## 6 Male 54 0 0 never 31.9 6.6
## 7 Male 79 0 0 former 31.2 5.8
## 8 Male 54 0 0 former 32.8 5
## 9 Male 38 0 0 never 55.6 6.5
## 10 Male 58 0 0 former 36.5 5.8
## # ℹ 7,435 more rows
## # ℹ 2 more variables: blood_glucose_level <dbl>, diabetes <dbl>male_data %>% filter(bmi >= 30, diabetes == 0) %>% tally()## # A tibble: 1 × 1
## n
## <int>
## 1 7445# the assumption is that overweight people are more likely to have diabetes. Below is the code and tally of WOMEN who are overweight in terms of bmi and DONT have diabetes
female_data %>% filter(bmi >= 30, diabetes == 0)## # A tibble: 11,852 × 9
## gender age hypertension heart_disease smoking_history bmi HbA1c_level
## <chr> <dbl> <dbl> <dbl> <chr> <dbl> <dbl>
## 1 Female 54 0 0 former 54.7 6
## 2 Female 78 0 0 former 36.0 5
## 3 Female 53 0 0 No Info 31.8 4
## 4 Female 34 0 0 never 56.4 6.2
## 5 Female 77 1 1 never 32.0 5
## 6 Female 27 0 0 not current 30.2 5.7
## 7 Female 37 0 0 No Info 30.5 5.7
## 8 Female 56 0 0 never 31.0 6.5
## 9 Female 44 0 0 never 37.4 5.7
## 10 Female 30 0 0 No Info 50.1 6
## # ℹ 11,842 more rows
## # ℹ 2 more variables: blood_glucose_level <dbl>, diabetes <dbl>female_data %>% filter(bmi >= 30, diabetes == 0) %>% tally()## # A tibble: 1 × 1
## n
## <int>
## 1 11852Note that the echo = FALSE parameter was added to the
code chunk to prevent printing of the R code that generated the
plot.
# in this section we want to display a stack bar showing how HbA1c levels (average blood sugar levels) can classify Males and Females in different categories such as Normal, Prediabetes, and Diabetes. From this plot we gain insight on how the distribution of Normal, Prediabetes, and Diabetes varies between both genders.
# first we load our libraries such as dplyr for data manipulation, ggplot2 for data visualization, and plotly to make our plot interactive.
library(dplyr)
library(ggplot2)
library(plotly)
# here we display our original dataset
diabetes_dataset## # A tibble: 100,000 × 9
## gender age hypertension heart_disease smoking_history bmi HbA1c_level
## <chr> <dbl> <dbl> <dbl> <chr> <dbl> <dbl>
## 1 Female 80 0 1 never 25.2 6.6
## 2 Female 54 0 0 No Info 27.3 6.6
## 3 Male 28 0 0 never 27.3 5.7
## 4 Female 36 0 0 current 23.4 5
## 5 Male 76 1 1 current 20.1 4.8
## 6 Female 20 0 0 never 27.3 6.6
## 7 Female 44 0 0 never 19.3 6.5
## 8 Female 79 0 0 No Info 23.9 5.7
## 9 Male 42 0 0 never 33.6 4.8
## 10 Female 32 0 0 never 27.3 5
## # ℹ 99,990 more rows
## # ℹ 2 more variables: blood_glucose_level <dbl>, diabetes <dbl># from our original data set we create a new dataset called HbA1c_by_gender. For our new dataset we want to keep only the Male and Female gender and exclude Other, the way to do this is by filtering our gender column to not equal to 'Other'. We also create a new variable called HbA1c_category with the use of our mutate function. We use the case_when function to classify our HbA1c_level column and return different categories such as 'Normal', 'Prediabetes', and 'Diabetes'.
HbA1c_by_gender <- diabetes_dataset %>% filter(gender != 'Other') %>%
mutate(HbA1c_category = case_when(
HbA1c_level < 5.7 ~ "Normal < 5.7%",
HbA1c_level >= 5.7 & HbA1c_level < 6.5 ~ "Prediabetes 5.7% - 6.4%",
HbA1c_level >= 6.5 ~ "Diabetes ≥ 6.5%",
TRUE ~ NA_character_ # handles unexpected cases that don’t match the defined conditions keeping missing values as NA
))
# here we print out our mutated dataset to make a comparison of the original.
HbA1c_by_gender## # A tibble: 99,982 × 10
## gender age hypertension heart_disease smoking_history bmi HbA1c_level
## <chr> <dbl> <dbl> <dbl> <chr> <dbl> <dbl>
## 1 Female 80 0 1 never 25.2 6.6
## 2 Female 54 0 0 No Info 27.3 6.6
## 3 Male 28 0 0 never 27.3 5.7
## 4 Female 36 0 0 current 23.4 5
## 5 Male 76 1 1 current 20.1 4.8
## 6 Female 20 0 0 never 27.3 6.6
## 7 Female 44 0 0 never 19.3 6.5
## 8 Female 79 0 0 No Info 23.9 5.7
## 9 Male 42 0 0 never 33.6 4.8
## 10 Female 32 0 0 never 27.3 5
## # ℹ 99,972 more rows
## # ℹ 3 more variables: blood_glucose_level <dbl>, diabetes <dbl>,
## # HbA1c_category <chr># in this section we want our stack bar to follow a certain order, we want Normal at the top, Prediabetes at the middle, and Diabetes at the bottom. To do this we will mutate our HbA1c_category column to an ordered categorical variable with the use of the factor function, the levels argument helps to set the order we want.
HbA1c_by_gender <- HbA1c_by_gender %>%
mutate(HbA1c_category = factor(HbA1c_category, levels = c("Normal < 5.7%", "Prediabetes 5.7% - 6.4%", "Diabetes ≥ 6.5%")))
# now we will plot a stack bar using ggplot. Before that we will change the name of our data set to 'Interactive_mode' that way we can easily incorporate our data set to ggplotly. The scale_fill_manual function is used to manually assign colors to the different categories.
interactive_mode <- HbA1c_by_gender %>% ggplot(aes(x = gender, fill = HbA1c_category)) +
geom_bar(position = "stack") +
scale_fill_manual(values = c("Normal < 5.7%" = "cornsilk2", "Prediabetes 5.7% - 6.4%" = "darkkhaki", "Diabetes ≥ 6.5%" = "darkgoldenrod")) +
labs(title = "Male vs. Female HbA1c Levels",
x = "Gender",
y = "Count",
fill = "HbA1c Category") +
theme_classic() +
theme(plot.title = element_text(hjust = 0.5)) # this code adjusts the tittle to the middle
# by using ggplotly we transform our ggplot into an interactive plot. When approaching the table we are able to see the count, gender, and HbA1c_category for any of the stacked bars.
ggplotly(interactive_mode)# here I'll leave extra info for you guys regarding the gender column of the original data set
diabetes_dataset %>% filter(gender == 'Female') %>% tally # 58,552 we have 17,122 more females than males in this data set## # A tibble: 1 × 1
## n
## <int>
## 1 58552diabetes_dataset %>% filter(gender == 'Male') %>% tally # 41,430 ## # A tibble: 1 × 1
## n
## <int>
## 1 41430diabetes_dataset %>% filter(gender == 'Other') %>% tally # 18 ## # A tibble: 1 × 1
## n
## <int>
## 1 18In the smoking data there are 6 unique values
The total amount of people who fall into each category is as follows;
There is quite a sizable amount of people in the No info category.
The total number of people in the dataset is 100000. To help clean up the data, we can filter ‘No Info’ people out. When we do that we get 64184.
# Figure out the unique categories of smoking history
unique(diabetes_dataset$smoking_history)## [1] "never" "No Info" "current" "former" "ever"
## [6] "not current"# Count amount of people who belong to each unique category
diabetes_dataset %>% group_by(smoking_history) %>% summarise(total_people = n())## # A tibble: 6 × 2
## smoking_history total_people
## <chr> <int>
## 1 No Info 35816
## 2 current 9286
## 3 ever 4004
## 4 former 9352
## 5 never 35095
## 6 not current 6447smoking_diabetes_dataset <- diabetes_dataset %>%
filter(smoking_history != 'No Info') %>%
group_by(smoking_history, diabetes) %>%
summarise(total = n())## `summarise()` has grouped output by 'smoking_history'. You can override using
## the `.groups` argument.Now we can graph the relationship between
print('hello world')## [1] "hello world"